Kyle Rozema (Washington University; Google Scholar), Does the Bar Exam Protect the Public?, 18 J. Empirical Legal Stud. 801 (2021):

I study the effects of requiring lawyers to pass the bar exam on whether they are later publicly disciplined for misconduct. In the 1980s, by abolishing what is known as a diploma privilege, four states began to require graduates from all law schools to pass the bar exam. My research design exploits these events to estimate the effect of the bar passage requirement on the share of lawyers publicly sanctioned by state discipline bodies. I find that during the first decade of their careers lawyers licensed without a bar passage requirement are publicly sanctioned at similar rates to lawyers licensed after passing a bar exam. Small differences do begin to emerge after a decade, however, and larger though still modest differences form after two decades.

Conclusion
This article studied the effects of requiring lawyers to pass a bar exam to obtain a law license. To do so, it exploited four events in the 1980s where states abolished their diploma privilege for graduates of a single law school within the state. The difference-in-differences research design assessed changes in public sanctions against graduates of these four law schools who obtained a law license in the years before and after the diploma privilege was abolished. Control groups were formed using lawyers in other states from the same geographic region that did not change their bar passage requirement over the same time period. Using this research design, I found evidence that lawyers licensed without a bar passage requirement received public sanctions at similar rates to lawyers licensed after passing the bar exam until roughly the second decade after receiving a law license. Within 10 years after obtaining a law license, between 0.8% and 2.0% of lawyers licensed after passing the bar exam had received public sanctions, compared to 1.7% of lawyers licensed without a bar passage requirement. Within 25 years after obtaining a law license, the bar passage requirement decreased the share of licensed lawyers who received a sanction from 5.1% to 3.9%. The annualized rate of this cumulative share is 0.9%, and the confidence intervals suggest that the effect is not greater than an annualized rate of 1.6%.

There are several limitations of the study. First, I do not have individual level data for lawyers who graduated from the four relevant law schools and who were not publicly sanctioned, so I estimate the number of lawyers in the treatment groups after the bar passage requirement was imposed using available information from the number of applicants and bar passage rates. Although I present robustness checks suggesting that these estimates do not materially affect the results, better estimates could be obtained if state bars or state discipline bodies release data on licensed lawyers. Second, I use lawyers in other states from the same geographic region as a control group, but the legal markets for graduates of the four law schools in the treatment group are likely to be structurally different than the legal market for lawyers practicing in large cities in the control states. Although the research design accounts for the difference in number of lawyers in the treatment and control groups, a better control group could be formed with data on the law schools that individual lawyers attended and the legal markets that they work in. Third, the states that abolished the diploma privilege that I study also changed one or more other licensing requirements around the same time, including requiring passage of an additional exam on ethics and changing rules whereby applicants with a felony conviction are categorically barred from admission. As a result, the estimates only capture the effects of the package of licensing changes around the time the diploma privilege was abolished. Finally, the estimates are based on changes in the diploma privilege in four small states in the 1980s. The policy changes studied here thus occurred in states with legal markets that might not be comparable to other legal markets, and the effects of a bar passage requirement in the states studied here could be different than the effects in states that have always required lawyers to pass the bar exam.

Future research should explore the appropriate role of the bar passage requirement and other quality control systems in preventing negative outcomes. There are many different systems that could potentially serve as a quality control measure through their gatekeeping role of the legal profession, including the American Bar Association’s certification of law schools, law schools’ admissions and grading decisions, and state licensing bodies’ character and fitness reviews. The legal malpractice system and the regulation of malpractice insurance could also serve as quality control measures. Future research should explore the extent to which these other quality control systems overlap with the bar passage requirement. For example, how much overlap is there between the set of lawyers who fail the bar exam at least once, the set of lawyers that state licensing bodies identify as having questionable moral character and fitness, and the set of lawyers who are eventually subjected to discipline? And how much overlap is there between the conduct that is subject to legal malpractice and the conduct that lawyers are disciplined for by state discipline bodies?39 Future research should also explore the extent to which the bar passage requirement interacts with other quality control systems. For example, given that the bar exam measures aptitude and skill, which are the same characteristics that drive law school admissions decisions, do law school admissions and grading decisions respond to changes in the bar passage requirement? And do malpractice insurance costs change after a change in the bar passage requirement? Understanding the extent that the bar passage requirement influences the decisions of the American Bar Association and law schools would help inform a longstanding debate about the roles that governing bodies, law schools, and courts do and should play in regulating the legal profession.